The need for high-speed internet and data communication for aircrafts in flight has been growing in the past decade. Broadband transmission schemes and systems between mobile carriers and satellite networks have been proposed and implemented to meet such aeronautical applications. Mobile satellite antennas for aeronautical applications need to be compact in size, light weight, high gain, and cause little interference to adjacent satellites. Waveguide-fed horn antennas and arrays have been the preferred choice for such applications, in particular for Ku band data transmission, because of their high aperture efficiency, low radio-frequency (RF) loss, and compact dimensions. As the communication frequency increases from Ku band to Ka band (up to 31 GHz) to meet the growing demand of high data-rate and low cost, antenna subsystems are facing more stringent requirements from both airliners and the regulatory commission.
For Ka band communication, air-born satellite antennas are required to use circular polarization (CP) in transmitting and receiving RF signals. In addition, the transmitting operation uses a higher frequency band (27.5-31 GHz) with right- or left-hand circular polarization (RHCP or LHCP), and the receiving operation occurs in a lower frequency band (17.7-21.2 GHz) with the opposite circular polarization. To generate CP RF fields with a horn antenna, two orthogonally linear-polarized (LP) wave fields with 90 degree phase difference are needed that propagate from the feed-waveguide to the horn aperture. To achieve this, a polarizer is usually required that converts single LP wave field into two orthogonal LP wave fields with required phase difference. Due to the high frequencies used and limited radome space, it's critical that proposed polarizers, waveguide feed networks, and horn antennas are as compact as possible, and can be easily integrated together to achieve a high array aperture efficiency.
Various techniques have been proposed for circular polarization in waveguides for general purpose, including capacitive or inductive iris loading in a square waveguide [1], a stepped septum [2], a linear-slope [3] or a curved-slope [4] septum in a square waveguide, and dielectric slabs in circular waveguides [5]. However, those techniques either work for only one relative narrow frequency band, or if capable of broadband or dual band, lack of a frequency or polarization diplexing function, or are large in size and cannot meet the limited space requirements for airborne Ka-band satellite antennas. There are two key issues that a circular polarizer needs to address to be able to use with aeronautical Ka-band satellite horn antennas, in particular, in array environment.
First, because of the requirements of minimum possible size and weight, a CP antenna or array must work as both the transmitter (RHCP) and receiver (LHCP), which means a circular polarizer should possess the polarization diplexing function to eliminate a separate diplexer. Septum based circular polarizers are also CP diplexing, due to their bifurcation nature, whereas, those polarizers based on irises, dielectric slabs, etc, need additional LP diplexers, such as an orthogonal mode transducer (OMT), which significantly increases the size and weight.
Second, because of the requirements of maximum antenna efficiency, a square aperture is preferred for CP horn antennas in array environments (circular aperture results in much lower array aperture efficiency due to the gaps between circular areas). So are the cross sections of the feeding waveguides and polarizers. For square waveguides, the dominant guided modes are TE10/TE01 (the two have same cut-off frequency), and existing septum based circular polarizers [2-4] were usually designed to include only these two dominant modes, which leads to a maximum bandwidth less than 25%. As was mentioned earlier for Ka band communications, the transmitting and receiving bands are quite far separated (the ratio of maximum frequency span to receiving band central frequency is 11.8:19.7) so that to entertain both bands, the waveguide polarizers have to include the TE11 mode, whose cut-off frequency is 1.4 times that for the TE10/TE01 modes and is below the transmitting band. The mode conversion in this case deteriorates the polarization performance in the higher band, and that needs to be addressed.
An approach [6] to avoid the higher order modes' presence was proposed primarily for reflector feeds that uses a dual waveguide structure in the polarizer design, where an inner waveguide polarizer and an out waveguide polarizer are coaxially arranged, each of these septum-based polarizers is operated to include only the dominant TE10/TE01 modes and used for carrying the linear-circular polarization conversion in only one frequency band. However, due to its coaxial dual waveguide structure nature, the proposed polarizer essentially prohibits the use of waveguide feed networks for dual frequency band signals, and therefore in horn array applications, it requires coaxial cable feed networks that would cause significant signal loss for the high frequency Ka band (30 GHz or higher). Besides, the plurality structures of the polarizer would result in a higher fabrication cost compared to single waveguide polarizers.
For waveguide feed networks, a compact patented design [7] has been proposed that consists of only E-plane tees and H-plane bends. However it requires a 90-deg twist at each element antenna port, which adds additional loss and blocks the implementation of two separate waveguide feed networks for transmitting and receiving signals respectively. A recent proposal of a waveguide antenna array for Ku band applications [8] uses a binary tree of E- and H-plane dividers. However, the waveguide network designed has taken up two thirds of the whole array thickness (dimension in the direction normal to the array aperture surface), resulting in shorter horn antennas with lower gain and efficiency.